The Western Union Company WU

Key data

The business

Western Union moves money across borders for the people the banking system serves last: migrants sending wages home, often in cash, picked up in cash at the other end. Two engines. The retail network, a large base of agent locations where someone hands dollars across a counter, is the old cash cow and it is shrinking. Branded digital, the app and website, is growing but is still the smaller share and grows slower than retail falls. The moat is real and narrow: the payout end, the trusted storefront in Oaxaca or Cebu where a grandmother collects the money, and a brand older than most of its customers. That moat is melting at the send end, where a phone undercuts the counter on price and speed. Every incumbent in a network business swears the switch away will be gradual, right up until it isn't. The August move to buy Intermex, a Latin-America-focused rival, is consolidation: buy volume and agents in the corridors that still pay, and cut cost.

The numbers

The trajectory is a slow bleed the market has watched for years.

The top line is roughly flat quarter to quarter, but earnings have halved: net income of $76.7M in Q2 2026 against $122.1M a year earlier, EPS of $0.24 against $0.37. On the adjusted figure the Street tracks, WU earned $0.31 then $0.25 the last two quarters against expectations near $0.42, two clean misses that took the stock from $10 toward $7.

Zoom out and the shape is unmistakable. Revenue has fallen every year since 2021, from $5.07B to $4.05B, about -5.4% a year compounded, and 2026 is tracking lower still. This is not a growth company; it is a cash machine running down. The saving grace is that the machine is capital-light: capex runs near 1% of revenue, so almost all operating cash converts to free cash, an 18% free-cash yield at today's price. The catch is the balance sheet. Long-term debt of $2.51B against $0.92B cash leaves net debt near 1.6 times EBITDA, and operating profit covers interest under two times, a thin cushion for a business whose revenue only goes down.

The market prices WU as an ice cube that melts to nothing. The variant is duration: if the decline settles at low single digits rather than accelerating, the cash returned between now and then is worth far more than 6 times earnings implies. The single print that settles it is the revenue decline rate, quarter by quarter.

Management

Insiders are net sellers, modestly: about $241,000 sold against $90,000 bought over the last year, plan status not disclosed on the sales, so read them as neither a scream nor a signal. The one buy of note, director Giovanni Angelini's $90,000 in November 2025, was partly reversed by a $57,000 sale five months later. What matters more is capital allocation, and here management is consistent: it repurchased $235M of stock in 2025 and pays a dividend that, at $7.18, yields about 13% and eats roughly three-quarters of earnings. The prior worry, that the dividend run-rate is the quarter the reason to own it breaks, held this quarter but only just: adjusted earnings of $0.31 still covered the payout, while GAAP earnings of $0.24 barely did.

How it fails or surprises you

The dividend gets cut (downside). The payout consumes about three-quarters of earnings, net debt sits at 1.6 times EBITDA, and interest is covered only 1.85 times. If adjusted EPS slips below the roughly $0.94 annual run-rate for two or three quarters, the board faces a choice, and a cut erases the one reason most holders are here. Watch quarterly EPS and operating cash flow against the $0.94.

The corridors decline faster (downside). The US excise tax on money sent abroad took effect in January 2026, and digital rivals keep taking the younger sender. If the -5% annual slide steepens toward -8% or worse, the duration bet fails and the ice cube melts on schedule. Watch retail versus branded-digital transaction growth in the next print.

Intermex works and the slide stops (right tail). Buy a rival cheap, strip cost, and stabilize revenue at low-single-digit decline while returning 13% a year and buying back shares at 6 times earnings, and the stock re-rates hard. The market pays nothing for this today. Watch for the decline rate narrowing alongside accretion guidance.

Closing thoughts

The Street knows what WU is: a deep-value, high-yield stock in visible secular decline. the case against it, that any yield on a shrinking business is a trap, is the consensus. The edge, if there is one, is narrow and it is about duration, not direction: everyone agrees WU shrinks, the disagreement is how fast and for how long the cash keeps coming. The left tail is fatter and nearer, a dividend cut on top of an accelerating decline, and you own a value trap that halves from here. The right tail, stabilization plus years of a 13% check and heavy buybacks at 6 times earnings, is real but pays out slowly and needs Intermex to deliver.

The bet is still that Western Union keeps handing you a dividend worth about 13% a year faster than its money-transfer business shrinks, and that the millions of people who walk into a shop to wire cash home keep doing it. What breaks it is dividend coverage, EPS and operating cash flow against the $0.94 payout, and the revenue decline rate. If both hold and the decline stays shallow, you are paid to wait; if either cracks, the reason to own it is gone.